What is it about?
This research investigates how different machine learning (ML) algorithms perform in the development of credit scorecards. It compares the effectiveness of these algorithms when using alternative data alone versus traditional credit bureau data in a retail bank setting.
Featured Image
Photo by PiggyBank on Unsplash
Why is it important?
This research finds that alternative data, including an applicant's social network default status, regional credit status, and local demographics, can influence creditworthiness. This suggests that alternative data is a valuable tool for assessing credit risk, particularly for borrowers who lack a traditional credit history. This could be instrumental for banks in expanding access to credit and promoting financial inclusion.
Perspectives
This research explores the potential of alternative data to improve credit scoring and expand access to credit for a wider population.
Rivalani Hlongwane
University of Cape Town
Read the Original
This page is a summary of: Enhancing credit scoring accuracy with a comprehensive evaluation of alternative data, PLoS ONE, May 2024, PLOS,
DOI: 10.1371/journal.pone.0303566.
You can read the full text:
Contributors
The following have contributed to this page
